Field of Science

The DaVinci stereogram

A few posts ago, I described how to make stereograms. At the end of the post, I showed a second type of stereogram, in which an illusionary white box appears to float in front of a background of Xs, and I promised to explain how that one was done.

This type of stereogram, discovered by Nakayama and colleagues, is called a "DaVinci stereogram" in honor of the famous artist/engineer who worked out the logic centuries ago (though he didn't, to my knowledge, consider building any stereograms).

The idea works like this: Look at an object (such as your computer monitor). Your left eye can
"see around" the left side of the object a bit more than can your right eye, while your right eye can see more of what is behind the object than can your left eye. It turns out that this information alone is sufficient to induce a perception of depth.

Consider that final stereogram (reproduced here). In both images, there is a white box in the center. However, the left image (the one presented to the right eye, if you use the divergence method) has four extra Xs on the right side of the box, while the right image (the one presented to the left eye) has two extra Xs on the left side of the box. This results in the perception of a white box floating above the background.


Anonymous said...

"However, the left image (the one presented to the right eye, if you use the divergence method)"

that would be the convergence method, not the divergence method.

josh said...

Is it? I've never actually been clear about these terms. Are we counting the convergence of the eyes or the images?

If somebody could diagram this, I'd actually find it very helpful.